Hi Pierre,
While the name is different, the MSE criterion is strictly equivalent
to the reduction of variance. The only difference is that we do not
divide by var{y|S} because this factor is the same for all splits and
all features, hence the maximizer is the same.
Cheers,
Gilles
On 24 February 2015 at 01:53, Pierre-Luc Bacon <pba...@cs.mcgill.ca> wrote:
> In the original Extra-Tree papers, the authors use the "relative variance
> reduction" (appendix A) for regression.
>
> The implementation in Scikit-Learn however suggests a different criterion:
> https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/tree/_tree.pyx#L836
>
> What was the rational behind this choice ?
>
> Thanks,
> Pierre-Luc
>
> ------------------------------------------------------------------------------
> Dive into the World of Parallel Programming The Go Parallel Website,
> sponsored
> by Intel and developed in partnership with Slashdot Media, is your hub for
> all
> things parallel software development, from weekly thought leadership blogs
> to
> news, videos, case studies, tutorials and more. Take a look and join the
> conversation now. http://goparallel.sourceforge.net/
> _______________________________________________
> Scikit-learn-general mailing list
> Scikit-learn-general@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
